Quality Check Skill
Use this skill when the user asks to run repository quality checks and summarize failures with actionable fixes.
Purpose
Execute the repository quality gates in a predictable order and return a concise report of:
- •what passed
- •what failed
- •likely root cause per failure
- •suggested fix per failure
Scope Rules
- •Keep changes minimal and focused on quality-check-related issues.
- •Do not add new dependencies unless the user explicitly asks.
- •Do not fix unrelated failures outside the user request.
- •Prefer project scripts from
package.jsonover ad-hoc commands.
Command Sequence
Run from repository root.
Core quality checks
- •
npm run style-check - •
npm run spell-check - •
npm run lint - •
npm run compile - •
npm run test
Optional validations (run when requested or relevant)
- •
npm run build(aggregated check path) - •
npm run check-manifests(if manifest changes are in scope)
Execution Behavior
- •Continue running remaining checks even if one check fails.
- •Capture per-command:
- •status (
pass/fail) - •exit code (if available)
- •short error summary
- •key file paths/lines from diagnostics when present
- •status (
- •Use platform shell commands:
- •Windows:
cmd /c "<command>" - •macOS:
zsh -lc "<command>"
- •Windows:
Reporting Format
Return a compact report with these sections:
- •Checks Run
- •ordered list of commands with pass/fail status
- •Failures Found
- •one bullet per distinct failure containing:
- •failing command
- •primary error message
- •likely root cause
- •suggested fix (specific and minimal)
- •one bullet per distinct failure containing:
- •Suggested Next Step
- •whether to re-run a single failing check or full suite
Suggested Fix Heuristics
- •Formatting errors: recommend
npm run stylethen re-runnpm run style-check. - •Lint errors: recommend targeted code fixes; use
npm run lint-fixonly when safe. - •Spelling errors: update source text or approved dictionary per repo conventions.
- •Compile errors: report TypeScript/Webpack diagnostics with file + symbol focus.
- •Test failures: distinguish test assertion failures vs. environment/setup issues.
Deliverables
- •Command-by-command quality check result summary.
- •Failure report with suggested fixes.
- •Clear rerun recommendation.